Computational stereo estimates 3D structure by analyzing visual changes between two or more passive images of a scene that are captured from different viewpoints. It is a key enabler for ubiquitous autonomous systems, large-scale surveying, virtual reality, and improved techniques for compression, tracking, and object recognition. The fact that computational stereo is an under-constrained inverse problem causes many challenges. Its computational and memory requirements are high. Typical heuristics and assumptions, used to constrain solutions or reduce computation, prevent treatment of key realities such as reflection, translucency, ambient lighting changes, or moving objects in the scene. As a result, a general solution is lacking. Stochastic models are common in computational stereo, but stochastic algorithms are severely under-represented. In this dissertation I present two stochastic algorithms and demonstrate their advantages over deterministic approaches. I first present the Quality-Efficient Stochastic Sampling (QUESS) approach. QUESS reduces the number of match quality function evaluations needed to estimate dense stereo correspondences. This facilitates the use of complex quality metrics or metrics that take unique values at non-integer disparities. QUESS is shown to outperform two competing approaches, and to have more attractive memory and scaling properties than approaches based on exhaustive sampling. I then present a second novel approach based on the Hough transform and extend it with distributed ray tracing (DRT). DRT is a stochastic anti-aliasing technique common to computer rendering but which has not been used in computational stereo. I demonstrate that the DRT-enhanced approach outperforms the unenhanced approach, a competing variation that uses re-accumulation in the Hough domain, and another baseline approach. DRT’s advantages are particularly strong for reduced image resolution and/or reduced accumulator matrix resolution. In support of this second approach, I develop two novel variations of the Hough transform that use DRT, and demonstrate that they outperform competing variations on a traditional line segment detection problem. I generalize these two examples to draw broader conclusions, suggest future work, and call for a deeper exploration by the community. Both practical and academic gaps in the state of the art can be reduced by a renewed exploration of stochastic computational stereo techniques. / text

This thesis examines the stereotypes of Taiwanese held by European students in Taiwan. One hundred and thirty-six European students approached by email and in person were administered a questionnaire composed of 34 questions. The results showed that Taiwanese people were in general considered as “hardworking”, “very friendly”, “superstitious” and “modest”. Eight hypotheses about knowledge about Taiwan and images of Taiwan held prior to arrival, Chinese proficiency, perceived similarity between people in students’ home countries and Taiwanese people, frequency and closeness of contacts with Taiwanese people, willingness to stay in Taiwan after graduation and closest accepted relation with Taiwanese person and their respective influence on the overall image were tested. Prior knowledge about Taiwan was found positively related to perceived similarity and closest accepted relation. Perceived similarity was found related to the number of declared Taiwanese friends and willingness to stay and along with the rise of the number of Taiwanese friends, the perceived similarity was greater and the students were more willing to stay in Taiwan. Also students who declared more Taiwanese friends accepted closer relations with Taiwanese people. Pre-arrival images, Chinese language proficiency, frequency of contacts were shown not to influence overall image of Taiwanese held by European students in Taiwan.

No description available.

This thesis presents an indoor obstacle avoidance system for car-like mobile robot. The system consists of stereo vision, map building, and path planning. Stereo vision is performed on stereo images to create a geometric map of the environment. A fast sparse stereo approach is employed. For different areas of the image there are different optimal values of disparity range. A multi-pass method to combine results at different disparity range is proposed. To reduce computational complexity the matching is limited to areas that are likely to generate useful data. The stereo vision system outputs a more complete disparity map. Abstract Map building involves converting the disparity map into map coordinates using triangulation and generating a list of obstacles. Occupancy grids are built to aid a hierarchical collision detection. The fast collision detection method is used by the path planner. Abstract A steering set path planner calculates a path that can be directly used by a car-like mobile robot. An adaptive approach using occupancy grid information is proposed to improve efficiency. Using a non-fixed steering set the path planner spends less computation time in areas away from obstacles. The path planner populates a discrete tree to generate a smooth path. Two tree population methods were trialled to execute the path planner. The methods are implemented and experimented on a real car-like mobile robot.

We report on experiments with a mobile robot using one vision process (forward motion vision) to calibrate another (stereo vision) without resorting to any external units of measurement. Both are calibrated to a velocity dependent coordinate system which is natural to the task of obstacle avoidance. The foundations of these algorithms, in a world of perfect measurement, are quite elementary. The contribution of this work is to make them noise tolerant while remaining simple computationally. Both the algorithms and the calibration procedure are easy to implement and have shallow computational depth, making them (1) run at reasonable speed on moderate uni-processors, (2) appear practical to run continuously, maintaining an up-to-the-second calibration on a mobile robot, and (3) appear to be good candidates for massively parallel implementations.

We describe a method to solve the stereo correspondence using controlled eye (or camera) movements. These eye movements essentially supply additional image frames which can be used to constrain the stereo matching. Because the eye movements are small, traditional methods of stereo with multiple frames will not work. We develop an alternative approach using a systematic analysis to define a probability distribution for the errors. Our matching strategy then matches the most probable points first, thereby reducing the ambiguity for the remaining matches. We demonstrate this algorithm with several examples.

A method has been developed to measure the perceived depth of computer generated images of simple solid objects. Computer graphic techniques allow for independent control of different depth queues (stereo, shading, and texture) and enable the investigator thereby to study psychophysically the interaction of modules for depth perception. Accumulation of information from shading and stereo and vetoing of depth from shading by edge information have been found. Cooperativity and other types of interactions are discussed. If intensity edges are missing, as in a smooth-shaded surface, the image intensities themselves could be used for stereo matching. The results are compared with computer vision algorithms for both single modules and their integration for 3D vision.

An important part of stereo vision consists of finding and matching points in two images which correspond to the same physical element in the scene. We show that zeros of curvature of curves are perspective projection invariants and can therefore be used to find corresponding points. They can be used to help solve the registration problem (Longuet-Higgins, 1982) and to obtain the correct depth when a curve enters the forbidden zone (Krol and van de Grind, 1982). They are also relevant to theories for representing image curves. We consider the stability of these zeros of curvature.

It is commonly assumed that the goal of stereovision is computing explicit 3D scene reconstructions. We show that very accurate camera calibration is needed to support this, and that such accurate calibration is difficult to achieve and maintain. We argue that for tasks like recognition, figure/ground separation is more important than 3D depth reconstruction, and demonstrate a stereo algorithm that supports figure/ground separation without 3D reconstruction.

A new noniterative approach to determine displacement vector fields with discontinuities is described. In order to overcome the limitations of current methods, the problem is regarded as a general modelling problem. Starting from a family of regularized estimates, by measuring the difference in description length the compatibility between different levels of regularization is determined. This gives local but noisy evidence of possible model boundaries at multiple scales. With the two constraints of continous lines of discontinuities and the spatial coincidence assumption consistent boundary evidence is found. Based on this combined evidence the model is updated, now describing homogeneous regions with sharp discontinuities.